"Clipper" Distortion Fades In - Advice Please

[iheartleaven]

This is kind of a weird one. I built a New Clipper, to this guy's schematic: "http://www.runoffgroove.com/newclipper.html" a while back, real simple circuit, only added an LED, and as far as I ever noticed, everything always worked fine.

Recently, however, I decided to install a 3PDT switch for full bypass, and a stereo in-jack for battery on/off. Ever since I made these additions, it has exhibited some strange behavior.

Basically: it works fine in bypass, but when you turn the distortion on, there's an audible moment of silence before the effect sort of "fades" in. After that, the effect is fine, and the transition back to a bypassed state sounds perfectly smooth. The pause upon switching to the effect is very consistent in its audibility and duration, so I doubt the problem is a loose solder joint.

My one theory is that something in the circuit is acting as a big capacitor, taking time to fill up before things get going. However, I've checked and re-checked my caps, and they seem to function correctly. Does anyone with more experience have any suggestions as to what could possibly be going on? I'm totally stumped.

[GibsonGM]

Did you do the 'ground the input of the effect' trick?  That might be causing it, by draining the power supply (?).  I'm not 100% sure on this one, about 95%, lol, but that 100uF cap in the PS is pretty hefty...   .0001F*22000 Ohms = 2.2 secs, the RC time constant - is that about the 'time lag'? If so, I'd suggest NOT grounding the input, but use the simple "always on" method where you're only switching the throughput to the effect and vice-versa....

***EDIT***
First, tho, make sure your switching jack is set up correctly, so that the ground line on the switch has nothing to do with the effect being on (it should be on as soon as the plug goes in the jack, check for voltages at opamps etc).   If you've accidentally got it switching ground on with the footswitch, you'll get that behavior, too....

[iheartleaven]

I had that thought about the input grounding as well -- unfortunately, I already tried un-grounding the input of the effect, and I even went so far as to take that 100uF out of the way altogether, but the problem remains. The time lag is closer to about one second... really wondering what's gone wrong here. 

Thanks for the input!